Intel Corporation

Qualcomm Incorporated

Nefeli Networks, Inc.

Intel Corproation

Kevin Roland Fall

Charles E Narad

Neil MacAvoy

Jerry J Hall

Pradip Shankar

Leonard M Rand

Thomas Stockhammer

Sylvia Ratnasamy

Michael G Luby

Charles Nung Lo

Daniele Di Proietto

David Naylor

Matthew Mussomele

Gordon Kent Walker

Nagaraju Naik

Carlos Marcelo Dias Pazos

Yinian Mao

Albert Christopher Torek

Kuo-Chun Lee

Jun Wang

Fatih Ulupinar

Nikolai Konrad Leung

Salman Abdul Baset

Lorenz Christoph Minder

Jack Shauh

Qiang Gao

Allan Douglas Knies

Gianluca Iannaccone

Maziar H Manesh

Chris Torek

Katerina Argyraki

Byung-Gon Chun

Norbert Egi

Mihai Dobrescu

Michael George Luby

A method for self-monitored universal scaling of software network functions involves receiving, at a switch of a network, one or more batches of data units. The network further includes one or more network function (NF) instances of an NF service, and a scaling controller. The switch transmits first data units to an NF instance of the NF service during a first time period. A first distribution associated with the NF instance is updated using the first transmitted data units. Upon determining that the updated distribution has changed such that a first measure of the first distribution is outside of a first confidence interval threshold, the first distribution is reinitialized. The switch transmits second data units to the NF instance during a second time period. The reinitialized first distribution is updated using the second transmitted data units to produce a second distribution associated with the NF instance.

Self-monitoring universal scaling controller for software network functions

A method for universal scaling of software network functions involves receiving, at a switch of a network, a batch of data units during a first period. The network further includes one or more network function (NF) instances of an NF service, and a scaling controller. The switch transmits one or more units of data during the first period to an NF instance of the NF service. An estimated maximum safe data unit rate is determined for the NF instance, and a representative safe data unit rate is determined for the NF service. A total number of data units designated to be received by the NF service during the first period is determined, and a total number of NF instances of the NF service to be provisioned in the network is determined at the scaling controller using the estimated total number of data units and the representative safe data unit rate.

Universal scaling controller for software network functions

Systems, methods, and devices of the various embodiments enable HTTP redirect messages to indicate content access policy information. In this manner, policy rules may be implemented based on the policy information in the HTTP redirect messages to control network congestion and/or quality of service ('QOS').

Method and apparatus for using HTTP redirection to mediate content access via policy execution

A proxy unit is configured to obtain mapping information that maps an identifier for media data to a resource location based on a service for retrieving the media data, wherein the service defines at least one of a plurality of types of transports for transporting the media data, receive a request for the media data from an application service client, determine whether the service is available, and, when the service is available, cause the application service client to receive the media data from a unit that receives the media data using the service from the resource location, based on the mapping information. In this manner, the application service client may receive media data from the unit (e.g., a middleware unit), which then receives the media data using a service that defines transport according to, e.g., broadcast or multicast transport, or another fashion (e.g., unicast) if the defined transport is unavailable.

Supporting transport diversity and time-shifted buffers for media streaming over a network

A method, an apparatus, and a computer program product for communicating data via the Transmission Control Protocol (TCP) are provided. The apparatus receives from a sender a plurality of data segments according to a sequential order, infers at least one missing data segment in the sequential order, and acknowledges to the sender receipt of a highest byte sequence number of a received data segment plus one based on the inference of the at least one missing data segment. The apparatus also adjusts a receive window based on at least one of: the inference of the at least one missing data segment, or a round-trip-time of a TCP connection, wherein the receive window indicates a buffer capacity for receiving data segments. The apparatus further advertises the adjusted receive window to the sender.

Avoiding unwanted TCP retransmissions using optimistic window adjustments

In one example, a client device for receiving information for streaming of media data includes a clock, and one or more processors configured to receive a media presentation description (MPD) for media content, wherein the MPD includes data indicative of wall-clock times at which the client device can retrieve data of the media content from a source device, and wherein the data indicates a synchronization method by which the client device is to synchronize the wall-clock times with the clock, synchronize the clock with the wall-clock times using the method indicated by the MPD, and request data of the media content from the source device using the synchronized clock. A server device or source device may form and provide such an MPD to a client device.

Live timing for dynamic adaptive streaming over HTTP (DASH)

A client device presents streaming media and includes a stream manager, a request accelerator, and a source component coupled to the stream manager and the request accelerator for determining which requests to make. A rate selection process can make rate decisions so that the buffer is filled when it is low, avoiding erratically changing rates and can choose the correct steady rate quickly. Multimedia download strategies can be used for HTTP that allow for accurate rate estimations, achieving link capacity even if network delays and packet loss rates are high, achieving timely delivery of the stream, and achieving relatively steady download rates with little short term variability. A receiver might use multiple HTTP connections, decompose media requests into smaller chunk requests, synchronize the connections using TCP flow control mechanisms, and request data in bursts. In addition, the receiver might use an HTTP pipelining process to keep the connections busy.

DASH client and receiver with playback rate selection

Data objects can be delivered over a network using a file delivery system and universal object delivery and template-based file delivery. This might be done by forming source data into a sequence of data objects represented by symbols in packets, sending those to receivers on request, wherein a transmitter obtains a template file delivery table with delivery metadata for the data objects, and constructing a first transmission object identifier for a data object based on a transmission object identifier construction rule described in the template file delivery table. A receiver might receive packets, extract a second transmission object identifier, associate encoded symbols comprising the received data packet with the data object if the first transmission object identifier and the second transmission object identifier identify the same data object, and recover, at least approximately, the source data for the data object based on the encoded symbols associated with the data object.

Universal object delivery and template-based file delivery

Embodiments enable HTTP servers to pass incomplete and/or corrupted files in response to file requests from clients. In the various embodiments, HTTP servers may be enabled to generate status codes identifying that an incomplete version of a file is being returned in response to a file request. In an embodiment, an HTTP server may be enabled to determine the ability of a client to handle incomplete versions of files.

Responding to hypertext transfer protocol (HTTP) requests

Systems, methods, and devices for dynamic resolution of content references for streaming media are described. In one innovative aspect, a device for obtaining a media stream is provided. The device includes a receiver configured to receive a location independent persistent identifier for the media stream. The device includes a resolution circuit configured to identify a location of the media stream based at least in part on the location independent persistent identifier. The device includes a media acquisition circuit configured to obtain the media stream based at least in part on the identified location. In some implementations, the resolution circuit is configured to identify a rewrite rule for the media stream and generate a rewritten identifier based on the identified rewrite rule and at least one of the location independent persistent identifier or the identified location.

Dynamic resolution of content references for streaming media

Generally, this disclosure describes a scalable cluster router that includes a plurality of server-class computers interconnected together to form a router. Each server may be configured to independently schedule switching of packets to reduce the switch speed requirements on a per server basis. Each server may include a scheduler that independently load balances packet flows across servers of the cluster. Router capacity may be incrementally scaled by adding more servers, and router capacity may be increased by load balancing techniques within individual servers.

Scalable cluster router

The present application describes a compiler of a network packet classification programming language that generates code for processors such as an application processor and a processing engine. The programming language includes a variety of instructions including an instruction to declare a network protocol and an instruction to specify a rule and at least one action to perform if the rule applies. A processor executing instructions generated by the compiler assigns values based on instructions to declare a network protocol and applies the rule instructions to received packets. The programming language may also include other instructions such as an instruction to search a set of values and identify whether an encapsulated packet header is present in a packet.

Compiler for computer programming language including instruction statements for handling network packets

Described herein are techniques to perform reassembly of a Transmission Control Protocol (TCP) data stream from payloads of TCP segments of a bidirectional TCP connection between a first TCP end-point operating at a first network device and a second TCP end-point operating at a second network device.

Reassembly of a transmission control protocol (TCP) data stream from payloads of TCP segments of a bidirectional TCP connection

The present invention relates to a general-purpose programmable packet-processing platform for accelerating network infrastructure applications which have been structured so as to separate the stages of classification and action. Network packet classification, execution of actions upon those packets, management of buffer flow, encryption services, and management of Network Interface Controllers are accelerated through the use of a multiplicity of specialized modules. A language interface is defined for specifying both stateless and stateful classification of packets and to associate actions with classification results in order to efficiently utilize these specialized modules.

Growing community of inventors

Bethel Park, PA, United States of America

Kevin Roland Fall